Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

blogs.nvidia.com

Post reply on HN
Print view View original

141–150 of 347 posts

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#141

Post by ghaff »

jefft255 wrote: post #123
dannyw wrote: post #90

Earlier quoted context omitted.

I'm sure Nvidia is paying a nice licensing fee to Ampere for the registered trademark. Otherwise it'd be sheer stupidity.

I don’t know if you’re being sarcastic, but trademarking a famous scientist’s name seems legally dubious at best. Is that allowed? I.e. Tesla the car company and nvidia Tesla’s GPUs; is there a gentleman’s agreement between these two or is such a name simply not trademarkable?

My understanding (IANAL) is that "Tesla" is trademarked by the company. [1] Note that doesn't mean no one else can ever use that name for anything. It does mean I'd probably get a cease and desist letter if I tried to go into the car business under the name Tesla. But I can probably get a trademark in other areas where there isn't a serious possibility of confusion. How close I'm willing to get to that line probably depends on how badly I want the name and how conservative the lawyers are.

[1] https://www.lexology.com/library/detail.aspx?g=e4eaa83d-47dc...

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#142

Post by thu2111 »

macksd wrote: post #71

The numbers for their SATURNV supercomputer are either untrue or absolutely staggering. 4.6 exaflops? #1 on the Top 500 list of supercomputers just barely passed 200 petaflops at peak performance. If you add up the entire list you only get 1.65 exaflops. And LINPACK isn't usually network-bound. How can this possibly be true?

AI only perf vs general perf.

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#143

Post by wongarsu »

jefft255 wrote: post #123
dannyw wrote: post #90

Earlier quoted context omitted.

I'm sure Nvidia is paying a nice licensing fee to Ampere for the registered trademark. Otherwise it'd be sheer stupidity.

I don’t know if you’re being sarcastic, but trademarking a famous scientist’s name seems legally dubious at best. Is that allowed? I.e. Tesla the car company and nvidia Tesla’s GPUs; is there a gentleman’s agreement between these two or is such a name simply not trademarkable?

In general a trademark is registered for a specific class (~industry). As long as the trademarks are for different classes and there is no risk of confusion, multiple companies/products can have the same name. For example Delta is the name of an airline, a computer company, a faucet company and a family of orbital rockets. There's little risk anybody would confuse these four, and they don't belong to the same class, so they can coexist just fine.

The US is quite generous in what you can trademark. In the EU for example you can't (as easily) trademark common words. But a quick search shows about 50 active EU trademarks on the word Tesla.

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#144

Post by thu2111 »

jamesblonde wrote: post #48
lowdose wrote: post #38

> DGX-A100 - The First HPC System With 140 Peta-OPs Compute Shipping Now For $199,000 Crazy pricing for a chip that basically started as a game solution. Anybody actually seen these Nvidia DGX systems in the wild?

We are working with 20 of them at a customer....

Can you talk about what sort of things you're doing, in general terms? Does the new generation DGX look like a worthwhile upgrade for you?

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#145

Post by chodeboy »

Fun fact: nVidia CEO Jensen Huang and AMD CEO Lisa Su are niece and uncle. They both are hardware engineers by trade and both wear leather jackets during presentations (although she took after him).

Must make for some interesting holiday family dinner talks.

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#146

Post by fluffything »

adev_ wrote: post #121
fluffything wrote: post #94

Earlier quoted context omitted.

> They make good hardware, but there's a lot of lock-in, and not a lot of transparency. This sounds like you'd like NVIDIA to open-source all their software. I see this type of request a lot, but I don't see it happening. NVIDIA's main competitive advantage over AMD and Intel is its software stack. AMD could release a 2x powerful GPGPU tomorrow for half the price and most current NVIDIA users wouldn't care because wh…

> AMD could release a 2x powerful GPGPU tomorrow for half the price and most current NVIDIA users wouldn't care because what good is that if you can't program it?. Correction: Nobody will be able to use the AMD hardware (outside of computer graphics) because everybody has been locked-in with CUDA on Nvidia. They can not even change even if they want to: it is pure madness to reprogram an entire GPGPU software stack e…

> Correction: Nobody will be able to use the AMD hardware (outside of computer graphics) because everybody has been locked-in with CUDA on Nvidia.

NVIDA open-sourced their CUDA implementation to the LLVM project 5 years ago, which is why clang can compile CUDA today, and why Intel and PGI have clang forks compiling CUDA to multi-threaded and vectorized x86-64 using OpenMP.

That you can't compile CUDA to AMD GPUs isn't NVIDIA's fault, it's AMD, for deciding to pursue OpenCL first, then HSA, and now HIP.

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#147

Post by jayd16 »

No one is talking about it but the mention of a focus on "datacenter computing" (in the video) with CUDA is interesting. In retrospect a GPU is basically running a map/reduce type workflow so its not that crazy after all.

Is this new or is CUDA already being used as a distributed language?

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#148

Post by taurath »

fluffything wrote: post #94
wegs wrote: post #23

Earlier quoted context omitted.

Yeah. That's been my general problem with adopting NVidia for anything. They make good hardware, but there's a lot of lock-in, and not a lot of transparency. That introduces business risk. I'm not in a position where I need GPGPU, but if there wasn't that risk, and generally there were mature, open standards, I'd definitely use it. The major breakpoint would be when libraries like Numpy do it natively, and better yet…

> They make good hardware, but there's a lot of lock-in, and not a lot of transparency. This sounds like you'd like NVIDIA to open-source all their software. I see this type of request a lot, but I don't see it happening. NVIDIA's main competitive advantage over AMD and Intel is its software stack. AMD could release a 2x powerful GPGPU tomorrow for half the price and most current NVIDIA users wouldn't care because wh…

This has literally been a back and forth argument since a 100 point post on slashdot was a groundbreaking event. I don't see it changing any time soon - honestly if anything on tech forums this argument frequently overshadows just how well NVIDIA is doing.

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#149

Post by the_pwner224 »

tyingq wrote: post #22
xvilka wrote: post #16

Probably even more closed than ever. They tend to become more and more restrictive with every new hardware generation. I wonder where their promised open source announcement they preannounced before.

I suspect some of it is driven by trying to keep the gaming/ai and desktop/server markets from overlapping. Market segmentation. If it were more open, that would be harder.

I wanted to do GPU PCI passthrough in a VM (run Linux host, then for gaming run a Windows VM with the GPU passed through to get good performance). Nvidia disabled this for their consumer GPUs; the Nvidia drivers in the Windows VM will block this from working. It was a purely software thing; there was no reason for this aside from nvidia wanting companies to pay more for the Quadro/etc. GPUs.

In addition to that, there's always the proprietary blob running in my kernel.

So a few months ago I bought an AMD card for my new computer.

Post reply on HN